A model is only useful if it is accurate on real data, explainable to the people who rely on it, and running where they can use it. That is how I build: from problem framing and data checks to evaluation, explanation and deployment.
What you get
- Prediction and classification (risk, churn, fraud, demand)
- Natural language processing, including African languages
- Computer vision on medical and other images
- Explainable results with SHAP and LIME
- Deployment as an app or API with Streamlit, FastAPI or Flask
Proof from real work
- OncoVision: breast cancer survival prediction (C-index 0.69)
- PhytopathAI: breast histology classification (85% on held-out images)
- A hybrid deep learning system for breast tumour classification
- 40% faster cyber-threat identification for a client
How it works
- You send a short note about what you need. It takes about a minute.
- I reply within 24 hours with questions or a first plan.
- We agree scope, timeline and price. Nothing is owed until you say yes.
Questions people ask
Can you build an AI model for my business?
Yes. Tell me the decision you want to improve and the data you have; I will tell you honestly whether machine learning is the right tool.
Do you deploy models?
Yes, as an app or an API, so your team can use the model without touching code.
Can you explain how a model makes its predictions?
Yes. I use explainability methods such as SHAP so results can be checked and trusted.
Work with Dominion Akinrotimi
Tell me what you need. I reply within 24 hours.
Other services: Hire a data scientist · Data analysis · Data engineering · Research support · Speaking · Career guidance · Training